Capra：适用于高分辨率IMAP分析的可扩展HEALPix原生强度重建
Capra: Scalable HEALPix-Native Intensity Reconstruction for High-Resolution IMAP Analyses
浏览论文内容
中文总结 AI 辅助
Capra是一款HEALPix原生管道，可从IBEX和IMAP任务的事件计数数据重建全天ENA强度图，生成两类产品，能减少伪影、验证不变性、具备线性强缩放，适合高分辨率IMAP重建。
中文摘要 AI 辅助
我们提出了Capra，一款用于从NASA星际边界探测器（IBEX）和星际测绘与加速探测器（IMAP）任务的事件计数数据重建全天高能中性原子（ENA）强度图的HEALPix原生管道。Capra生成两类产品：（1）基线视轴分配图；（2）HEALPix网格上的可选平滑图。这两类产品可生成平滑、物理解释明确的图，具备一致的分辨率变化和参考框架变换。平滑后的Capra图可恢复大尺度ENA形态，同时减少采样导致的斑驳和条状伪影。与Theseus重建的相对比较显示，二者大尺度结构大体一致，差异源于重建方法的不同，而非主要形态差异。我们验证了计数数量、速率和强度在坐标变换下的 tessellation（网格划分）不变性（保持性），并在代表性数据集上展示了其数值收敛性可达HEALPix分辨率Nside=64。我们还表明，基于线程的并行性性能呈现近似线性强缩放，内存使用受限，可实现高效的高分辨率重建，使Capra非常适合高分辨率IMAP重建。
英文摘要
We present Capra, a HEALPix-native pipeline for reconstructing all-sky energetic neutral atom (ENA) intensity maps from the NASA Interstellar Boundary Explorer (IBEX) and Interstellar Mapping and Acceleration (IMAP) missions' event-counting data. Capra produces two products: (1) a baseline boresight-assigned map, and (2) an optional smoothed map on the HEALPix grid. This produces smooth, physically interpretable maps with consistent resolution changes and reference-frame transformations. The smoothed Capra maps recover the large-scale ENA morphology while reducing sampling-driven mottling and strip-like artifacts. Relative comparison with the Theseus reconstruction shows broadly consistent large-scale structure, with the discrepancies reflecting differences in reconstruction methods, rather than major morphological discrepancies. We verify invariance (preservation) of count numbers, rates, and intensities across tessellations under coordinate transformations, and demonstrate numerical convergence up to HEALPix resolution Nside = 64 for a representative dataset. We show that thread-based parallelism performance exhibits approximately linear strong scaling, with bounded memory usage, enabling efficient high-resolution reconstruction and making Capra well-suited for high-resolution IMAP reconstruction.
